US 6934869 B1
A method and apparatus for eliminating dead zone in a phase locked loop with binary quantized detectors are described. Dead zone can be eliminated by changing the threshold used to quantize the cross point sample. A quantized cross point sample is integrated in order to set a new threshold. The integration may be performed during data transitions to eliminate threshold drift during long sequences where no transitions occur.
1. A phase locked loop having a binary quantized phase detector, comprising:
a first means for storing a reference threshold value;
a comparator having a first input for receiving a first input signal and a second input for receiving the reference threshold value, the comparator comparing the first input signal with the reference threshold value to generate an output; and
a set threshold block, comprising:
a second means for generating a constant value for level shifting the output from the comparator to produce an output whose value is 0; and
an integrator, coupled to the second means, for locating a cross point in a data-eye diagram of a data sequence, having an input for receiving the output from the comparator and an output for generating a new threshold value;
a third means for filtering the output of the second means, having a first input for receiving the quantized output and an output coupled to the integrator; and
a fourth means for determining if two values are not equal, having a first input for receiving a first input signal, a second input for receiving a second input signal, and an output coupled to a second input of the third means.
2. The phase locked loop of
3. A method for eliminating dead zone in a binary quantized phase detector, comprising:
referencing a reference data sample to locate a center of a first eye in a data-eye diagram;
receiving a first signal for a first data sample in the first center of the first eye;
receiving a second signal for a second data sample in a wandering interval caused by a dead zone;
using the reference data sample to locate a center of a second eye in the data-eye diagram;
receiving a third signal for a third data sample in the center of the second eye;
applying a first comparison to a value of the second signal and a reference threshold value;
applying a second comparison to a value of the first signal and a first reference constant;
applying a third comparison to the value of the third signal and a second reference constant; and
applying a function to a result of the second comparison and the third comparison, determining if the two comparison results are equal.
4. The method of
5. The method of
6. The method of
7. The method of
8. The method of
9. The method of
10. The method of
11. The method of
12. The method of
13. The method of
1. Field of Invention
The invention relates generally to phase locked loops, particularly to the elimination of dead zone for phase locked loops using binary quantized phase detectors.
2. Description of Related Art
Clock recovery circuits (CRC) are often used in communication systems and other electronic systems to synchronize a local clock to an external system clock. Currently, a phase locked loop (PLL) is the standard approach to constructing a CRC.
The phase and frequency detector 1 compares a reference signal denoted 7 to a feedback signal denoted 11 in order to produce a phase error signal denoted 9. The phase error signal 9 is then filtered through the low pass filter 3 and subsequently input to the VCO 5. The VCO 5 generates a signal 11 with a frequency controlled by the filtered phase error signal. The output 11 is fed back into the phase and frequency detector 1. If the two frequencies for the signals 7 and 11 do not equal, the filtered phase error signal would cause the VCO 5 to shift to the frequency of the reference signal 7. When the shift is completed, the output of the VCO 5 is used as the synchronized signal.
Several types of phase locked loops have been developed using phase detectors such as Alexander phase detector and Hogge phase detector. Particularly, the Alexander phase detector has been widely adopted due to its ease of implementation.
A CRC measures the clock phase and aligns it to the reference clock to minimize bit-error-rate, and the optimum sampling instant is at the center of the data-eye.
However, due to duty cycle distortion, the cross point 13 in
The present invention provides the method and apparatus for eliminating dead zone in a phase lock loop using a binary quantized phase detector. The present invention allows a binary quantized phase detector to locate a cross point in a data-eye diagram by gradually shifting a threshold value towards the threshold at which a cross point occurs.
In a first embodiment of the present invention, a binary quantized phase detector first receives a signal sampled in a wandering interval caused by duty-cycle distortion, and compares the value of the signal to a threshold value. The binary result of the comparison is then level shifted and subsequently integrated. The output of the integration is set as the new threshold and the process to locate the cross point restarts with the new threshold.
In a second embodiment of the present invention, the integration step takes place only when the data sequence transitions.
The accompanying drawings that are incorporated in and form a part of this specification illustrate embodiments of the invention and together with the description, serve to explain the principles of the invention:
The first embodiment shown in
Moreover, the scale factors such as 0.0001 in
As an example, the cross point shown in
Moreover, the output of the integrator block 4 is delayed one cycle in a delay block 6 and then used as the new threshold input 39 for the next comparison in the comparator 37, and the process to locate the cross point restarts with the new threshold 39. For the example shown in
In addition to the integration and shifting to find a cross point in a data-eye diagram, the second embodiment performs the integration step only when data transitions are present in order to eliminate threshold drift during long periods of data sequences without transitions.
The values 18 and 20 are binary results of comparing 0 to sample values taken at data-eye centers, with respect to the location of the wandering interval sample 16. If the values 18 and 20 differ, indicating a transition in the data-eye diagram, the multiplier 10 multiplies 1 with the quantized value 24, allowing the integration to proceed. Conversely, if the values 18 and 20 are equal, indicating a data sequence without transitions, the multiplier 10 multiplies 0 with the quantized value 24 and effectively eliminates the integration step.
Furthermore, since the samples 18 and 20 are located with reference to the location at which sample 16 is taken, the accuracy of the locations for the samples 18 and 20 is improved as the phase detector gradually shifts the amplitude at which it samples 16 towards the desired cross point.
As shown in
As the sample signal 16 is processed, a third sample taken at the center of a second eye is received in the circuit, and subsequently compared to 0 as illustrated by blocks 49 and 51. The resulting signal 20, whose value is either 0 or 1, is the output of the block 51. Subsequently, the signals 22, 18, and 20 are inputs to the circuit 41, whose components are explicitly illustrated in
As shown by blocks 41 and 37, the output 39 of the circuit 200 is the new threshold input for comparison with the wandering interval sample 16. Moreover, the comparison process continues with all three sample-points as the phase detector adjusts the sampling threshold through the integration performed in the circuit 41. Since the wandering interval sample 16 is used as a reference to locate the data-eye centers 18 and 20, as the phase detector adjusts the sampling amplitude to locate the cross point, it consequently adjusts the locations at which the samples 18 and 20 are taken and effectively allows the phase detector to self-center.
As illustrated in
The circuit of
A delay is then inserted to retime the Alexander phase detector in step 71. While the signal B is processed, two other signals A and C are sampled in steps 75 and 77 at the center of the eyes before and after B, using B as a reference mid-point location.
Both samples A and C are subsequently compared to 0 in steps 79 and 81. If the value of either signal is less than 0 (zero), the signal is set to 0 (steps 83, 87); otherwise, the signal is set to 1 (steps 85, 89). Step 91 then applies an XOR function to the resulting signals A and C to determine if a data transition has occurred in the interval between the two sample points.
Step 91 shows that a temporary variable T holds the result of the XOR function; if a transition exists in the data sequence, T equals 1; else T equals 0. T is then multiplied with the value of B in step 73 to determine whether integrating the value of B is necessary.
If the output O of step 73 is 0 (zero), the integrating step is overlooked, and the phase detector restarts the sampling process for all three signals. If the output O of step 73 is 1, the value of B is integrated in step 93, and the result is set as the new threshold for the phase detector (step 95).
Although the invention has been described in connection with several embodiments, it is understood that this invention is not limited to the embodiments disclosed, but is capable of various modifications that would be apparent to a person skilled in the art.
For example, the Alexander phase detector and the XOR function shown in the figures are intended as illustrations, other binary quantized phase detectors and means to determine if two values are equal may be used in conjunction with the present invention without departing from the spirits of the present invention. Moreover, the multiplication block 10 in
It is understood that common practice such as representing binary values with “0” and “1” or TRUE and FALSE may be implemented by various means without departing from the desired representation.
Temporary variables shown in
Furthermore, constants such as 0.5 used in
The foregoing descriptions of specific embodiments of the present invention have been presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed, and obviously many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and its practical application, to thereby enable others skilled in the arts to best utilize the invention and various embodiments with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims appended hereto and their equivalents.